SKAdNetwork industry benchmarks and best practices: 3 actionable strategies for improving attribution

Introduction

Measuring ad campaign performance on iOS is complex — even with the improvements with SKAdNetwork (SKAN) 4.0. Marketers often bring up the same challenges: missing conversion data, delays in data output, and discrepancies across measurement frameworks. If you feel like you're trying to solve a puzzle with missing pieces, you're not alone.

In part, this is because the adoption of SKAN 4.0 has been slow and steady with marketers and inconsistent with partners. Internal data from AppsFlyer shows:

• 45% of top-tier customers hold SKAN4 conversion schema and 26% of the total postbacks received are signed with version 4

• 80% of Moloco's traffic is signed with SKAN 4.0
• Other networks are spottier, with a few of the largest networks seeing less than 1% of their ads sign with SKAN 4.0 

Meanwhile, Moloco's internal data highlights a promising shift: the number of marketers looking at both mobile measurement partners (MMPs) and SKAN metrics has more than doubled, and those planning to adopt SKAN increased by 2.5x. While industry-wide adoption remains a work in progress, leaders recognize SKAN's potential to provide actionable insights when applied strategically.

Two of those experts — Priya Bhasin, Director of Product at Moloco and former Apple Search Ads leader, and Gil Or, Product Manager at AppsFlyer — recently co-hosted a webinar to share how marketers can find success with SKAN. During their discussion, they highlighted three proven strategies for optimizing measurement, campaign structures, and conversion schema to drive incremental improvements in iOS ad performance.

‍

‍

1. Adopt a hybrid measurement strategy and focus on overall performance — not individual metrics 

Different measurement approaches come with distinct performance analysis and discrepancies. MMP data is more granular and real-time while SKAN attribution is aggregated and delayed. 

Gil and Priya recommended adopting a hybrid measurement strategy to take advantage of what each framework offers. Effectively leveraging this strategy requires a deep understanding of possible discrepancies, such as:

• MMPs typically use a 7-day lookback window for click-based attribution, providing a broader view of user acquisition.
• SKAN deploys a unique fidelity score system, prioritizing StoreKit-rendered ads over view-through attributions, regardless of timing. 

MMP data provides the impression-level insights needed for real-time optimization, while SKAN data offers privacy-compliant validation and additional campaign success signals. This makes a hybrid approach especially valuable to performance-focused networks and DSPs like Moloco.

"It makes more sense to interpret each attribution method performance individually and prioritize the broader performance trend over isolated metrics. This approach helps in understanding overall campaign effectiveness without getting caught up in differences between attribution models." — Priya Bhasin 

2. Plan for minimum performance thresholds and data delays

Marketers regularly share their frustration at seeing a high rate of "NULL" values in their campaign data, effectively leaving them blind to user behavior. During the Q&A portion of the webinar, one attendee described spending over $60,000 on iOS campaigns but still seeing null SKAN data. 

Priya and Gil explained that high NULL conversion rates occur when campaigns don't meet SKAN’s “privacy thresholds” — and how marketers can respond:

• It’s assumed the “privacy threshold” dropped from 120 daily installs (per campaign) in SKAN 3.0 to around 20 in SKAN 4.0. 
• Marketers should keep targeting 100-150 installs to play it safe, since SKAN 4.0 isn't supported by all networks.
• To meet these thresholds, it's best to consolidate multiple small campaigns running simultaneously into one.

Campaign performance also requires patience and consistency. Campaigns should run for at least one month to allow a media source's algorithms to learn and improve. Additionally, whenever you update SKAN setups or conversion schema, expect three to four days of unreliable data as the system adjusts. 

"Meeting the privacy threshold is a frequent challenge that often results in missing conversions and high NULL rates. This also contributes to higher CPAs. You can overcome this challenge by consolidating several smaller campaigns into fewer high-volumes." — Priya Bhasin
‍

3. Optimize conversion schema to align measurement with business and performance goals 

Getting useful data from SKAN requires setting up a strategic conversion schema that maps the right events to track towards a specific goal — while also giving your ad partners the data they need for optimization. This can be tricky for marketers to get right, since there isn't a one-size-fits-all approach to conversion schema. 

Gil and Priya describe how different app categories require different approaches:

• Apps with in-app purchases perform best with a hybrid schema measuring both revenue and key events (like purchases and CPA goals).
• Ad-monetized apps typically focus exclusively on revenue measurement.
• Subscription-based apps often prioritize event measurement, since revenue generation happens outside SKAN's measurement window.

AppsFlyer provides schema templates and revenue recommendations based on industry benchmarks — because it's easy to lose sight of the forest for the trees, especially when it comes to campaign optimization.

"Marketers also need to consider that many networks, including the major ones, use SKAN signals for optimization. So if you're optimizing for value but aren't measuring revenue in your SKAN conversion schema, the networks won't have the data they need to provide you with strong performance. Don't forget to measure the key elements that are the base of what you're optimizing for." — Gil Or

Staying ahead of iOS attribution

Between ongoing SKAN updates and the forthcoming impact of AdAttributionKit (AAK), Apple will continue evolving its privacy-centric attribution. Marketers should work with partners who deeply understand and support these frameworks.

Moloco's machine learning capabilities and experience optimizing campaigns across MMP and SKAN data help marketers continually improve iOS ad performance. Connect with our team of experts to learn more. 

Reach out today
‍